ANALYTICAL

BIOCHEMISTRY
Analytical Biochemistry 331 (2004) 370–375
www.elsevier.com/locate/yabio

Colorimetric ferrozine-based assay for the quantitation

of iron in cultured cells
Jan Riemer,a,b Hans Hermann Hoepken,a,b Hania Czerwinska,b
Stephen R. Robinson,b and Ralf Dringena,b,¤
a
Interfakultäres Institut für Biochemie der Universität Tübingen, Hoppe-Seyler-Str. 4, D-72076, Tübingen, Germany
b
Department of Psychology, Monash University, Wellington Rd., Clayton, Victoria, 3800, Australia

Received 3 March 2004

Available online 6 May 2004

Abstract

The ferrozine-based colorimetric assay described here permits the quantitation of iron in cultured cells in amounts ranging
between 0.2 and 30 nmol. Ferrous and ferric iron were detected equally well by the assay and the accuracy was unaVected by other
divalent metal cations. This colorimetric assay was used to study iron accumulation in brain astrocytes that had been cultured in 24-
well dishes. Iron complexed to cellular proteins was made accessible to ferrozine by treatment of cell lysates with acidic KMnO4 solu-
tion. The basal amounts of iron in untreated astrocyte cultures were approximately 10 nmol iron per mg protein. Incubation of the
cells with ferric ammonium citrate caused the total cellular iron content to increase in a concentration-dependent manner. The esti-
mates of cellular iron content that were obtained with the ferrozine-based assay did not diVer from those determined by atomic
absorption spectroscopy. The colorimetric assay described here provides a sensitive, cheap, and reliable method for the quantitation
of intracellular iron and for the investigation of iron accumulation in cultured cells.
 2004 Elsevier Inc. All rights reserved.

Keywords: Assay; Astrocytes; Brain; Cultured cells; Iron; Transport; Uptake

As an essential component of heme groups and iron
sulfur clusters, iron is important for electron transport
in the respiratory chain and for various enzymatic
reactions. When present in excess, however, iron can
harm biological systems since in redox-active form it
catalyzes the generation of highly reactive oxygen spe-
cies [1]. Since both iron deWciency and overload impair
cellular functions [1–3], the quantitation of iron in cells
and extracellular Xuids is of considerable interest. The
quantitation of iron in biological samples can be
achieved with a variety of methods including atomic
absorption spectroscopy (AAS)1 [4,5], paramagnetic res-
onance spectroscopy [6], densitometric analysis [7], and
colorimetric quantitation [8–10].
¤
Corresponding author. Fax: +49-7071-295360.
E-mail address: ralf.dringen@uni-tuebingen.de (R. Dringen).
1
Abbreviations used: AAS, atomic absorption spectroscopy;
DMEM, Dulbecco's modiWed Eagle's medium; FCS, Fetal calf serum;
FAC, ferric ammonium citrate; PBS, phosphate-buVered saline.
Among the compounds used for colorimetric iron
quantitation, ferrozine is acknowledged as an eVective
chelator of ferrous iron and has been used for the deter-
mination of iron in biological samples [8,11]. Ferrozine
binds ferrous iron, but not ferric iron, into a complex
that absorbs strongly at 550 nm [8]. After uptake into
cells, iron ions are quickly stored in ferritin or incorpo-
rated into protein complexes. Such complexed iron is not
accessible for quantitation by colorimetric assays. To
quantify the total iron content of cells (low-molecular-
weight iron plus protein-bound iron) the iron must Wrst
be released from proteins.
For brain cells, particularly astrocytes, the mecha-
nisms involved in iron accumulation are not well
understood. Astrocytes in culture have been reported
to take up iron via transferrin-dependent and trans-
ferrin-independent pathways [5,12,13]. The latter
pathway in particular is worthy of further investiga-
tion, but research progress has been slowed by the need

0003-2697/$ - see front matter  2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.ab.2004.03.049
 J. Riemer et al. / Analytical Biochemistry 331 (2004) 370–375
for access to expensive and specialized equipment that
is required for AAS or uptake studies of radioactive
iron.
Here, we describe an optimized method for the
quantitation of total iron in cultured cells that is based
on the colorimetric ferrozine method [8]. The original
method was improved and adapted for the quantita-
tion of total cellular iron in conXuent astrocyte cultures
that had been grown in 24-well dishes. This simple and
reliable assay is suYciently sensitive to determine the
basal iron content in untreated cultures and it can
accurately quantify iron accumulation in iron-treated
cells. Since the assay requires only a microtiter plate
reader, it will be useful for the investigation of cellular
iron accumulation in laboratories that lack the tech-
nology required for radioactive measurements and
AAS.
Materials and methods
Chemicals
The assay described here was developed, reWned, and
extensively characterized in Melbourne (Australia).
These data were conWrmed and extended with AAS in
Tübingen (Germany). Therefore, the Australian and
German sources of the chemicals are listed here. Dul-
becco's modiWed Eagle's medium (DMEM) was
obtained from Gibco (Invitrogen, Melbourne) and from
Life Technologies (Eggenstein, Germany). Fetal calf
serum (FCS) was from CSL (Melbourne) and from
Serva (Heidelberg, Germany). Streptomycin sulfate and
penicillin G were from Sigma (Melbourne) and from
Serva (Heidelberg, Germany). L-Ascorbic acid sodium
salt, bovine serum albumin, ferrozine (3-(2-pyridyl)-5,6-
bis(phenyl sulfonic acid)-1,2,4-triazine), and neocupro-
ine (2,9-dimethyl(1,10-phenanthroline)) were obtained
from Sigma (Melbourne or Deisenhofen, Germany).
Ferric ammonium citrate (FAC) was purchased from
ICN Biomedicals (Ohio, USA) and from Fluka (Neu-
Ulm, Germany). The 24-well cell culture dishes and 96-
well microtiter plates were from Nunc and Greiner
Bio-one (Germany).
Astrocyte cultures
Astrocyte-rich primary cultures derived from the
brains of newborn Wistar rats were prepared and main-
tained as described previously [14]. The cells were seeded
in 24-well dishes (300,000 viable cells per well in 2 mL)
and incubated in culture medium (90% DMEM, 10%
FCS, 20 U/mL penicillin G, and 20
g/mL streptomycin
sulfate). The medium was renewed every 7th day. All
results were obtained from 14- to 21-day-old cultures.
The majority of cells in the astrocyte-rich primary
371
cultures can be immunostained for the astrocyte marker
glial Wbrillary acidic protein [14,15].
Incubation of cells with ferric ammonium citrate
Astrocyte cultures in 24-well dishes were washed
twice with 2 mL of DMEM and incubated for 24 h with
500
L DMEM containing iron as FAC in the concen-
trations indicated. The iron content of the available
FAC preparations can vary between batches. The iron
content of the batch (Fluka, Germany) used for the
experiments illustrated in Figs. 1–4 contained 0.91 §
0.04 mol iron per mol FAC (n D 8) as quantiWed by AAS.
Iron accumulation was terminated by washing the cells
twice with 2 mL of ice-cold phosphate-buVered saline
(PBS; 10 mM potassium phosphate buVer, pH 7.4, con-
taining 150 mM NaCl). After completely removing the
PBS, the cells were frozen and the dishes were stored at
¡20 °C. Cells were lysed with 200
L of 50 mM NaOH
for 2 h on a shaker in a humidiWed atmosphere. Aliquots
of these lysates were used for the iron and protein deter-
minations.
To exclude the possibility that iron was unspeciWcally
attached to the cells in astrocyte cultures after FAC
treatment rather then taken up into the cells, the FAC-
treated cultures were washed twice with 2 mL PBS that
contained the iron chelator deferoxamine in a Wnal con-
centration of 1 mM. This treatment did not lower the
Fig. 1. Absorbance of the Fe2+–ferrozine complex formed with increas-
ing concentration of the standard FeCl3. The amount of iron indicated
on the abscissa was contained in a volume of 280
L in wells of a
microtiter plate. The data represent means § SD of triplicates of a rep-
resentative experiment. The SDs are smaller than the symbols repre-
senting the mean values. The increase in absorbance was linear
between 1 and 30 nmol (A) and between 0.2 and 1 nmol (B) of FeCl3.
For both concentration ranges the correlation coeYcients were 0.999.
372 J. Riemer et al. / Analytical Biochemistry 331 (2004) 370–375
Fig. 2. Absorbance of the Fe2+–ferrozine complex that is formed in the
presence of various ferric and ferrous iron salts. The data represent
means § SD of triplicates of a representative experiment.
Fig. 3. EVects of divalent metal ions on the detection of iron by its fer-
rozine complex. FeCl3 (0, 5, or 20 nmol) was mixed with 50 nmol of the
indicated metal salts and the absorbance of the Fe2+–ferrozine
complex was measured at 550 nm. The data represent means § SD of
triplicates of a representative experiment.
amount of detectable iron in the cell samples, demon-
strating that the iron detected was not accessible to an
extracellular iron chelator. In addition, the intracellular
accumulation of astrocytes after FAC treatment was
conWrmed by histo- chemical Perl staining (data not
shown).
Fig. 4. Iron content of astrocyte cultures after incubation with FAC.
The cells were incubated for 24 h in serum-free DMEM with FAC at
the concentrations indicated. The cells were lysed in 200
L of NaOH.
(A) Aliquots of the lysates were transferred to Eppendorf tubes and the
complexed iron was released by treatment with acidic KMnO4 solution
before the total iron content was determined by the ferrozine method;
(B) The iron content of the lysates was determined by AAS; (C) The
cell lysates were treated with acidic KMnO4 in cell culture wells and the
total iron content of the lysates was determined by the ferrozine
method. The data represent means § SD of nine wells derived from
three independently prepared 15- to 18-day-old astrocyte cultures. The
mean protein content of the cultures was 111 § 19
g per well.
Quantitation of intracellular iron by the colorimetric
ferrozine assay
Aliquots (100
L) of cell lysates were placed in Eppen-
dorf tubes and mixed with 100
L of 10 mM HCl (the
solvent of the iron standard FeCl3), and 100
L of the
iron-releasing reagent (a freshly mixed solution of equal
volumes of 1.4 M HCl and 4.5% (w/v) KMnO4 in H2O).
These mixtures were incubated for 2 h at 60 °C within a
fume hood, since chlorine gas is produced during the
reaction [8]. The HCl/KMnO4 pretreatment is suYcient
to release iron quantitatively from proteins, including the
iron-storage protein ferritin [16] and heme proteins such
as hemoglobin [17]. The HCl/KMnO4-mediated digestion
of iron-containing proteins is essential for iron
quantitation of heme proteins. For example, treatment of
hemoglobin with ferrozine does not lead to the recovery
of any Fe2+–ferrozine complex if the protein has not been
pretreated with acidic permanganate solution [17].
After the mixtures had cooled to room temperature,
30
L of the iron-detection reagent (6.5 mM ferrozine,
6.5 mM neocuproine, 2.5 M ammonium acetate, and 1 M
ascorbic acid dissolved in water) was added to each tube.
After 30 min, 280
L of the solution in each tube was
transferred into a well of a 96-well plate and the
 J. Riemer et al. / Analytical Biochemistry 331 (2004) 370–375
absorbance was measured at 550 nm on a microplate
reader. The iron content of the sample was calculated by
comparing its absorbance to that of a range of standard
concentrations of equal volume that had been prepared
in a way similar to that of the sample (mixture of
100
L of FeCl3 standards (0–300
M) in 10 mM HCl,
100
L 50 mM NaOH, 100
L releasing reagent, and
30
L detection reagent). The intracellular iron concen-
tration determined for each well of a cell culture was
normalized against the protein content of that well.
If determination of the protein content of individual
wells was not required, the assay was performed directly
in 24-well dishes. Frozen cells were lysed by application
to each well of 200
L 50 mM NaOH, 200
L 10 mM
HCl, and 200
L iron-releasing reagent. The dishes were
sealed with foil and incubated for 2 h at 60 °C, after
which 60
L of the detection reagent was added. After
further incubation for 30 min at room temperature,
280
L of the mixture was transferred into a well of a 96-
well plate and its absorbance measured at 550 nm and
compared to the absorbance of equally treated FeCl3
standards. The intracellular iron concentration deter-
mined for the well was normalized against the protein
content of replicate wells.
Quantitation of intracellular iron by atomic absorption
spectroscopy
Aliquots of the 50 mM NaOH lysates of astrocyte cul-
tures were used to quantify the total amount of iron by
AAS using a Perkin–Elmer 400 spectrometer as
described previously [5]. The intracellular iron concen-
tration of each well was normalized against the protein
content for that well.
Determination of protein content
The protein content of astrocyte cultures was deter-
mined according to the method of Lowry et al. [18] using
bovine serum albumin as the standard.
Determination of cell viability
Cell viability was analyzed by determining the activity
of lactate dehydrogenase in the incubation medium
using the microtiter plate assay described previously [19].
Incubation of cultured astrocytes with FAC in concen-
trations of up to 100
M did not not substantially com-
promise the viability of the cells compared to that of the
controls (0
M FAC) (data not shown).
Presentation of data
The experiments shown in the Wgures were carried out
at least three times with comparable results. The data are
presented as means § SD of values obtained from repre-
373
sentative experiments or from experiments performed on
three independently prepared cell cultures.
Results and discussion
A ferrozine-based colorimetric assay was tested for its
potential to quantify the total iron content of cultured
astrocytes. Ferrozine forms a complex with ferrous iron
that strongly absorbs light at 550 nm [8]. After applica-
tion of the ferrozine and ascorbate-containing reaction
mixture to FeCl3 solutions, the absorbance at 550 nm
rapidly increased to maximal values within 10 min and
then remained stable for at least 60 min (data not
shown). In the range between 0.2 and 30 nmol iron the
absorbance of the Fe2+–ferrozine complex was propor-
tional to the iron concentration (Fig. 1). The slopes of
the standard curves calculated from the data for the two
ranges of iron amounts analyzed in Figs. 1A and B are
almost identical. The mean extinction coeYcient at
550 nm calculated from 18 calibration curves was
26.98 § 0.96 mM¡1 cm¡1. This value is almost identical to
the extinction coeYcient of 27.9 mM¡1 cm¡1 that was
recently described for Fe2+–ferrozine at 562 nm [20].
Since untreated astrocyte cultures contain about 10 nmol
iron per mg protein [5], and conXuent astrocyte cultures
comprise 70–200
g of protein per well of a 24-well dish,
the ferrozine-based iron assay is sensitive enough to
determine the iron content of a single well.
We have previously shown that FAC is a soluble
source of iron for cells in the rat brain and for primary
astrocytes in culture [5,21]. To determine whether the
ferrozine-based assay is able to detect iron in solutions
of FAC and of other ferric and ferrous iron salts,
varying concentrations of ferrous ammonium sulfate,
ferrous sulfate, ferric chloride, and FAC were added to
the reaction mixture and the absorbance of the Fe2+–
ferrozine complex at 550 nm was determined. Almost
identical absorbance readings were recorded at each
amount of iron applied (5, 10, or 20 nmol) for the four
iron salts investigated (Fig. 2). These data demonstrate
that ferric and ferrous iron are equally well detected by
the assay, that iron in FAC can be quantiWed by the fer-
rozine-based assay, and that the ascorbate in the
detection reagent quantitatively reduced ferric iron to
ferrous iron.
To investigate whether the ferrozine-based quantita-
tion of iron can be perturbed by other divalent cations,
50 nmol of various divalent metal ions were added to 0,
5, or 20 nmol FeCl3 and the absorbance at 550 nm was
measured. Although present in excess, Mn2+, Mg2+, Ni2+,
and Zn2+ did not alter the absorbance of the Fe2+–ferro-
zine complex at 550 nm (Fig. 3), indicating that these
metal ions do not interfere with iron quantitation. By
contrast, the absorbance at 550 nm of 0 or 5 nmol iron
was slightly increased by application of 50 nmol Cu2+
374 J. Riemer et al. / Analytical Biochemistry 331 (2004) 370–375
(Fig. 3). This elevated absorbance was not prevented by
increasing the concentration of the copper chelator
neocuproine in the reaction mixture (data not shown).
However, the increased absorbance caused by excess
copper was not evident in solutions of 20 nmol iron.
Thus, copper is likely to be a confounding variable only
when present at concentrations that are at least 10-fold
higher than that of iron. Since iron concentrations are
severalfold higher than copper concentrations in cul-
tured astrocytes [22] and brain tissue [23], copper ions
are unlikely to confound the ferrozine-based quantita-
tion of iron in cultured brain cells.
The ferrozine-based iron assay was used to study the
iron content of astrocyte cultures as a model system. To
determine total intracellular iron, the iron complexed to
proteins was released by treating lysates with acidic
KMnO4 solution [8]. Incubation of cell lysates with HCl/
KMnO4 for 2 h at 60 °C produced the maximal detection
of iron in the samples. Lower incubation temperatures
or shorter incubation periods at 60 °C yielded lower esti-
mates for total iron (data not shown).
The basal total iron content of conXuent astrocyte
cultures was found to be 10.2 § 2.9 nmol iron per mg
protein (n D 27 from six experiments). This value
obtained by the ferrozine assay is closely comparable to
estimates previously obtained for astrocyte cultures by
AAS (9.3 § 1.2 nmol/mg; [5]).
The iron content of astrocyte cultures was also deter-
mined after treating the cells with FAC. After incubation
for 24 h with concentrations of up to 100
M FAC, the
total iron content of the cultures was found to have
increased in proportion to the concentration of added
iron (Fig. 4). After treating the cell lysates with acidic
KMnO4 solution to release complexed iron, aliquots
were measured in parallel by the ferrozine-based assay
and by AAS. Values obtained by both methods were
almost identical (Fig. 4), demonstrating the reliability of
the ferrozine-based method for quantitation of intracel-
lular iron in FAC-treated cells. To further simplify the
ferrozine-based assay, iron release and the formation of
the Fe2+–ferrozine complex were performed directly in
the wells of 24-well dishes. This treatment resulted in
iron values that did not diVer signiWcantly from those
obtained by AAS or by ferrozine assays that had been
performed in Eppendorf tubes (Fig. 4).
Cells that had been incubated for 24 h with 10, 25, 50,
and 100
M FAC in serum-free medium contained 2-, 6-,
11-, and 21-fold more iron, respectively, than cells that
had been incubated without iron (Fig. 4). Cells accumu-
lated up to 65% of the applied iron within 24 h (Table 1),
demonstrating their capacity to accumulate iron
eYciently even in the presence of low micromolar con-
centrations of FAC. Under the conditions used in our
study, transferrin was not present. Thus, our data con-
Wrm the existence of a transferrin-independent mecha-
nism of iron uptake in astrocytes [12,13].
Table 1
Iron accumulation by astrocyte cultures during incubation with FAC
Applied Cellular iron Accumulated iron
iron (nmol) content (nmol/well)
(nmol) (% of applied iron)
0 1.4 § 0.3 0 0
4.5 2.7 § 0.4 1.3 38
11.3 8.8 § 1.5 6.4 65
22.5 16.0 § 1.4 14.6 65
45.3 30.0 § 2.3 28.6 63
The cultures were incubated for 24 h in 0.5 mL DMEM containing
FAC. The indicated amounts of applied iron were obtained by correct-
ing for the iron content of the FAC batch used. The total cellular iron
content of the cultures was determined by the ferrozine method after
treatment of the cell lysates with acidic KMnO4 in the cell culture
wells. The data correspond to those presented in Fig. 4 (condition C).
The data represent means § SD of nine wells derived from three exper-
iments performed on 15- to 18-day-old astrocyte cultures. The mean
protein content of the cultures was 111 § 19
g per well.
In summary, the optimized ferrozine-based colori-
metric assay provides a reliable, fast, and cheap method
for the measurement of total intracellular iron concen-
trations of cells cultured in 24-well dishes. The use of a
microtiter plate reader enables the iron content of a large
number of samples to be measured simultaneously,
which is advantageous when conducting investigations
of iron uptake by cultured cells.
Acknowledgments
Much of the research in this paper was supported by
a Senior Research Fellowship awarded to R.D. by Neu-
roSciences Victoria. We thank the Department of Psy-
chology, Monash University for Wnancial support,
Professor U. Weser and Dr. H.J. Hartmann (Tübingen)
for giving us the opportunity to use their atomic absorp-
tion spectrometer, and Dr. G. Bishop (Monash Univer-
sity) for helpful comments on the manuscript.
References
[1] R.R. Crichton, S. Wilmet, R. Legssyer, R.J. Ward, Molecular and
cellular mechanisms of iron homeostasis and toxicity in mamma-
lian cells, J. Inorg. Biochem. 91 (2002) 9–18.
[2] J.L. Beard, J.R. Connor, Iron status and neural functioning, Annu.
Rev. Nutr. 23 (2003) 41–58.
[3] J.R. Burdo, J.R. Connor, Brain iron uptake and homeostatic
mechanisms: an overview, Biometals 16 (2003) 63–75.
[4] W. Slavin, Atomic absorption spectrometry, Methods Enzymol.
158 (1988) 117–145.
[5] H.H. Hoepken, T. Korten, S.R. Robinson, R. Dringen, Iron accu-
mulation, iron-mediated toxicity and altered levels of ferritin and
transferrin receptor in cultured astrocytes during incubation with
ferric ammonium citrate, J. Neurochem. 88 (2004) 1194–1202.
[6] A.N. Woodmansee, J.A. Imlay, Quantitation of intracellular free
iron by electron paramagnetic resonance spectroscopy, Methods
Enzymol. 349 (2002) 3–9.
[7] S.M. LeVine, M.J. Wulser, S.G. Lynch, Iron quantiWcation in cere-
brospinal Xuid, Anal. Biochem. 265 (1998) 74–78.
 J. Riemer et al. / Analytical Biochemistry 331 (2004) 370–375
[8] W.W. Fish, Rapid colorimetric micromethod for the quantitation
of complexed iron in biological samples, Methods Enzymol. 158
(1988) 357–364.
[9] C. Gay, J. Collins, J.M. Gebicki, Determination of iron in solu-
tions with the ferric–xylenol orange complex, Anal. Biochem. 273
(1999) 143–148.
[10] A. Huberman, C. Perez, Nonheme iron determination, Anal. Bio-
chem. 307 (2002) 375–378.
[11] F. Ceriotti, G. Ceriotti, Improved direct speciWc determination of
serum iron and total iron-binding capacity, Clin. Chem. 26 (1980)
327–331.
[12] Z.M. Qian, Q.K. Liao, Y. To, Y. Ke, Y.K. Tsoi, G.F. Wang, K.P. Ho,
Transferrin-bound and transferrin free iron uptake by cultured rat
astrocytes, Cell Mol. Biol. (Noisy-le-grand) 46 (2000) 541–548.
[13] S.Y. Jeong, S. David, Glycosylphosphatidylinositol-anchored
ceruloplasmin is required for iron eZux from cells in the
central nervous system, J. Biol. Chem. 278 (2003) 27144–
27148.
[14] B. Hamprecht, F. LöZer, Primary glial cultures as a model
for studying hormone action, Methods Enzymol. 109 (1985) 341–
345.
[15] J.M. Gutterer, R. Dringen, J. Hirrlinger, B. Hamprecht, PuriWcation
of glutathione reductase from bovine brain, generation of an antise-
rum, and immunocytochemical localization of the enzyme in neural
cells, J. Neurochem. 73 (1999) 1422–1430.
375
[16] M.E. May, W.W. Fish, The UV and visible sprectral properties of
ferritin, Arch. Biochem. Biophys. 190 (1978) 720–725.
[17] S.S. Panter, Release of iron from hemoglobin, Methods Enzymol.
231 (1994) 502–514.
[18] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein
measurement with the Folin phenol reagent, J. Biol. Chem. 193
(1951) 265–275.
[19] R. Dringen, L. Kussmaul, B. Hamprecht, DetoxiWcation of exoge-
nous hydrogen peroxide and organic hydroperoxides by cultured
astroglial cells assessed by microtiter plate assay, Brain Res. Brain
Res. Protoc. 2 (1998) 223–228.
[20] B.S. Berlett, R.L. Levine, P.B. Chock, M. Chevion, E.R. Stadtman,
Antioxidant activity of ferrozine-iron-amino acid complexes,
Proc. Natl. Acad. Sci. USA 98 (2001) 451–456.
[21] G.M. Bishop, S.R. Robinson, Quantitative analysis of cell death
and ferritin expression in response to cortical iron: implications
for hypoxia-ischemia and stroke, Brain Res. 907 (2001) 175–
187.
[22] G. Tholey, M. Ledig, P. Mandel, L. Sargentini, A.H. Frivold, M.
Leroy, A.A. Grippo, F.C. Wedler, Concentrations of physiologi-
cally important metal ions in glial cells cultured from chick cere-
bral cortex, Neurochem. Res. 13 (1988) 45–50.
[23] M.A. Lovell, J.D. Robertson, W.J. Teesdale, J.L. Campbell, W.R.
Markesbery, Copper, iron and zinc in Alzheimer's disease senile
plaques, J. Neurol. Sci. 158 (1998) 47–52.